(Topic ID: 261887)

explain MPU flashing


By oldschoolbob

4 days ago



Topic Stats

  • 24 posts
  • 6 Pinsiders participating
  • Latest reply 17 hours ago by oldschoolbob
  • Topic is favorited by 1 Pinsider

You

Linked Games

No games have been linked to this topic.

    Topic Gallery

    There have been 8 images uploaded to this topic. (View topic image gallery).

    S20200216_0004 (resized).jpg
    S20200216_0003 (resized).jpg
    IMG_4204a (resized).jpg
    IMG_4210 (resized).JPG
    IMG_4011 (resized).JPG
    IMG_4219 (resized).JPG
    IMG_4204 (resized).JPG
    MPU-Bally35_U11-PortA.jpg

    #1 4 days ago

    I'm working on a Bally -35. I get the flicker - then the LED goes on then off (first flash) - then the LED goes on then off (second flash) - then the LED goes on then off (third flash) - then on and off (forth flash) - then the LED stays on. Wouldn't that mean the PIA at U11 has a problem? Or does it mean the 555 timer is not turning off the LED?

    I replaced all the sockets including the U11 socket. The board was DOA when I got it.

    Thanks

    Bob

    #2 4 days ago
    Quoted from oldschoolbob:

    Then the LED stays on. Wouldn't that mean the PIA at U11 has a problem?

    The 5th flash was incomplete, so yes.

    Check for solder bridges on the socket pins. Check to make sure the legs of the PIA chips are clean and not tarnished (you can clean with a pink rubber eraser). Swap the PIAs around to see if there is a change in behavior between the chips. If yes, the chip could be bad (and if you have a neoloch tester, you could test it). If there's no change in behavior, check for continuity issues from the chip's legs through the socket and onto the PCB.

    http://pinwiki.com/wiki/index.php?title=Bally/Stern#Fourth_and_Fifth_flashes

    #3 3 days ago

    Thanks ForceFlow, I checked the PIA in my Neoloch tester and it tested good. But I changed it anyway. Still no fifth flash. I'll check the socket for continuity tonight. I'm sure that's where the problem is.

    Bob

    #4 3 days ago
    Quoted from oldschoolbob:

    Wouldn't that mean the PIA at U11 has a problem?

    The LED is connected to U11. U11 has failed the internal register tests leaving the LED stuck on.
    This internal test can fail when one of the Port A pins (PA0 - PA7 on pins 2 - 9 respectively) has an external short circuit - it usually happens in one of the bypass capacitors on one of the signals near a pin header.

    Do a resistance check on those U11 pins with respect to ground looking for any with much lower resistance. If you find any with lower resistance than the others then unsolder the respective bypass capacitor on that signal to see if it's resistive (shorted). You can boot the board without that capacitor.

    MPU-Bally35_U11-PortA.jpg

    #5 3 days ago

    Thanks Quench, I think your diagnosis is spot on. I lost a solder pad on pin 4. I tried to repair it by using a machined pin socket and soldering on the top side. I probably caused a short to the adjacent trace. I didn’t have any shop time today but I’ll check it as soon as I get back to the shop. (The dark spots in the photos are sharpie marks.)

    IMG_4204 (resized).JPG

    IMG_4219 (resized).JPG

    On a brighter note – I got my ROM programmer and I “THINK” I burned my first EPROM tonight. It’s Leon’s test ROM. I haven’t used it yet so I don’t know for sure if it works. As always I had a devil of a time loading the driver for the programmer but it seems to work fine now.

    Thanks

    Bob

    #6 3 days ago

    If during PIA test the diagnostic LED flash stays illuminated that is a fail for that test. Treat it as if the flash was missing. Usually flash stay on = defective PIA chip itself.

    Theory of operation describes each flash.
    http://arcarc.xmission.com/Pinball/PDF%20Pinball%20Misc/Bally%20Theory%20of%20Operation.pdf

    #7 3 days ago
    Quoted from oldschoolbob:

    I lost a solder pad on pin 4.

    You're still losing some solder pads (or was that from someones previous work?) - I think we might have discussed this before but care to tell us your method of desoldering/removing the IC sockets and your equipment?

    #8 2 days ago

    Thanks Andrew, I guess I should call that a half-flash. I've changed the PIA chips three times. (all three tested good on the tester) I think the problem is with the new IC socket I installed.

    Quench, this time I followed your example. I removed all the IC sockets. I pried off all the plastic covers then heated each pin and removed the pin as the solder melted. (but I didn't pull them out with my fingers) This is the soldering station I have (set at 550* F.) :

    https://www.circuitspecialists.com/csi-premier-75w-Soldering-Station.html

    But this still leaves solder in the holes. To remove the remaining solder I use a desoldering station similar to this (mine is no longer available).

    https://www.circuitspecialists.com/Compact_Desoldering_System.html

    IMG_4011 (resized).JPG

    This is where I lose the solder pads. The only pads I lose are the pads without a trace connected.

    This is a Bally board and I seem to have more problems with Bally boards. There is no temperature reading on the desoldering tool - only a potentiometer without any markings. I guess I had it set at about 80 percent. Maybe I have it set too hot.

    When I removed the header pins I used the same method (one pin at a time). Then removed the solder left in the holes with the desoldering tool. This time I set the heat to about 60 percent. That worked much better.

    Once I get this board fixed I have another Bally board to repair. I'll try less heat next time.

    Thanks

    Bob

    #9 2 days ago

    Ok, I think the key issue here might be that you're having trouble melting the old solder which seems more common with the Bally MPU boards and it's causing you to put excessive pressure and extended heat on the pads. The white spots you see on the fibreglass PCB next to the pads on that second picture above are indicative of this.

    After pulling the pins, I add a touch of solder to the pads so solder flows and then desolder the holes.

    Maybe next time instead of using my time expensive method of pulling the socket cover/pins one at a time, just add a touch of solder to each of the socket pins so you get the solder flowing then desolder like you would previously had and *gently* remove the socket. Note, when desoldering don't put pressure on the pads, just bring the desoldering gun back a touch so all its contact is against the pin/solder. Swirl the gun around the pin till you feel the pin has become loose in the molten solder then pull the trigger. Again, don't swirl with the gun against the pad.
    If then the socket doesn't want to come out, carefully inspect each pin by moving them with something pointy. The pins that move freely are desoldered, the pins that are still stiff need to be redone - add a touch of solder and desolder again.
    When all the pins are loose the socket will easily come out (I know easier said than done..)

    I tend to have my desoldering gun set at high temperature but everyone has a different opinion about this.

    #10 2 days ago

    I add solder often before 'sucking' out old components. Getting a good flow with fresh solder makes the removal easier. Even when you pull the leg or component and there is some solder remaining, add some new and then suck it clean.

    I run my temp at 650. I do chage it occasionally, but not often. I also have a traditional iron and have no clue what temp the iron runs. But experience tells me when I'm too long (hot) or too short (cold). I also find tips important. Small pointy tips for board work and slightly larger for thicker holes. Medium when soldering wires. And so on. The time it takes to heat with the tips also makes a difference.

    #11 2 days ago

    Always learning. Dave is right - tips are important. I usually use the larger pointed tip for most work but when I worked on that SMD board I switched to a small chisel point. That worked really well there. I also used it to remove the sockets and headers. But I switched back to the pointed tip for installing parts.

    I also learned from the SMD board that flux helps. Before I removed anything I hit it with a little flux and melt the solder before turning the board over to remove the parts. The down side of the flux is it leaves a sticky residue that difficult to remove. I seldom add solder because it seems so counterproductive - but I've seen where it does help.

    I think my biggest mistake is using too much pressure on the pads with the desoldering gun. That's one thing I know I'm doing.

    I don't mind the time expensive method. I use to use the other method of desoldering first but this way seems much safer to me. Besides I have nothing better to do.

    By the way Quench, how's the weather where you are? It's been in the single digits (*F) here lately. (another reason I'm not getting much shop time - too cold down there). If you have better weather maybe I'll grab that other Bally board and stop by and you can show me your technique on repairing boards.

    Thanks guys.

    #12 2 days ago
    Quoted from Skidave:

    I add solder often before 'sucking' out old components. Getting a good flow with fresh solder makes the removal easier.

    Indeed it does. The other thing is keeping the desoldering gun capture cylinder and filters regularly clean.

    Quoted from oldschoolbob:

    I think my biggest mistake is using too much pressure on the pads with the desoldering gun.

    And that's likely because the old solder is playing difficult to melt. Adding new solder beforehand will help a lot.

    Quoted from oldschoolbob:

    By the way Quench, how's the weather where you are? It's been in the single digits (*F) here lately.

    Ouch, that's cold! So I shouldn't be complaining about the 60°F we had today..

    #13 2 days ago
    Quoted from oldschoolbob:

    I think my biggest mistake is using too much pressure on the pads with the desoldering gun. That's one thing I know I'm doing.

    Rubbing the desoldering gun tip against the board is a good way to make the pad fall off.

    Something I see people do when using a desoldering gun is they pull the trigger too early. You gotta let the solder melt all the way through the hole to the top side before pulling the trigger or else you get a half desoldered hole.

    Also tip... If your hole did not desolder 100% clean on the first try add more solder. It will not desolder clean until you do so.

    Another tip... On bally board desoldering the black open frame sockets with a single leaf (RN sockets i think they are) like to pull tight and the leaf sticks to the component side of the MPU. When desoldering those sockets I use the gun tip to carefully push/hammer out the contact away from the component side which stops them from wanting to stick to the component side. Then they just fall out or you can pick them off with your fingers with no force. I actually find those RN sockets harder to desolder than the close from orange/brown or AUGAT ones because of that effect.

    #14 2 days ago

    I always use the pull the socket frame technique at this point, and I'm the guilty one of heating the pin from the top and pulling it out (quickly, obviously) with my fingers.

    Afterwards I use desoldering braid to clear the holes, since my Hakko 808 needs to be rebuilt or something - it just stopped working effectively. I always add solder, and I'm really strange vs. other people in that I like to solder/desolder around 780-860 or so. More heat for less time. I haven't lifted a pad off in years.

    #15 2 days ago
    Quoted from slochar:

    I'm the guilty one of heating the pin from the top and pulling it out (quickly, obviously) with my fingers.

    Heh, I heat from the bottom and pull out with my fingernails The pins that are originally soldered with their leads bent over and don't want to come straight out make it interesting on the fingertips!

    I currently set my desoldering gun to around 680°F

    #16 2 days ago
    Quoted from slochar:

    I always use the pull the socket frame technique at this point, and I'm the guilty one of heating the pin from the top and pulling it out (quickly, obviously) with my fingers.
    Afterwards I use desoldering braid to clear the holes, since my Hakko 808 needs to be rebuilt or something - it just stopped working effectively. I always add solder, and I'm really strange vs. other people in that I like to solder/desolder around 780-860 or so. More heat for less time. I haven't lifted a pad off in years.

    I do 750f usually. I like big tips too =D. With the higher temp and big tip when you have a bunch of solder joints lined up in a row like an IC socket you can just drag the iron across the pins and as long as you keep adding the right amount of solder surface tension fills in the solder where it needs to go. It goes a lot quicker that way.

    Thankfully I desolder very rarely anymore unless I make a big mistake. I have grown to hate doing desoldering from the time I spent fixing originals.

    I am guilty of heating a hole and banging the board on the desk to get the solder to fly out if I just need to clear one hole. A blast of air will do it too.

    #17 1 day ago

    This has been a very interesting thread. I know I picked up a bunch of tips and tricks. I really enjoy rebuilding these old boards except when I lose a solder pad. I'm getting better but not 100 percent yet. Learning these tips and practice maybe I'll make it.

    Another thing I'd like to hear about is what you guys do to repair a lost pad. What is your method? How do you stitch the connection?

    Bob

    #18 1 day ago

    I had a little shop time today (a little warmer today). I was certain I had a short or open circuit somewhere so first I traced out all the pins in the missing pad areas. Everything checked out fine. Then I removed U11 and looked close under the socket - I even looked with my USB magnifier and held the board up to a bright light to view through it. I still found nothing. When I traced pin 2 (PA0) I noticed it was connected to U20 and U20 has been replaced before. I checked my parts box and found a new U20. Replacing it made no difference.

    I didn't have time (or patience) to do any other tests. Tomorrow I do the resistance check that Quench suggested back in post 4.

    Thanks

    Bob

    IMG_4210 (resized).JPG
    #19 1 day ago
    Quoted from oldschoolbob:

    Another thing I'd like to hear about is what you guys do to repair a lost pad. What is your method? How do you stitch the connection?

    I use machined IC sockets and solder the top side with my pointy tip iron. Sometimes I'll bend the offcut of a thin component lead into an oval shape and solder it to the leg so it looks like a solder pad.
    When the ground trace along the bottom of those MPU boards has been eaten away by corrosion, I use copper tape across the corroded trace and tin solder it. You can't tell it's there.

    #20 19 hours ago
    Quoted from Quench:

    Do a resistance check on those U11 pins with respect to ground looking for any with much lower resistance. If you find any with lower resistance than the others then unsolder the respective bypass capacitor on that signal to see if it's resistive (shorted). You can boot the board without that capacitor.

    I checked the resistance on pin 2 through pin 9. All showed about 7.4 K except pin 8 showed 4.9 K. Pin 8 is connected to C65.
    I checked resistance of C64 = 7.3 K. Then C65 = 4.9 K. Then C66 = 7.4 K. I pulled out the ground side of C65 and tried to boot - it still locks on at five flashes. With one leg of C65 disconnected I checked its resistance - it shows open (O.L). I checked the capacitance of C65 and it shows 0.49 nF.

    I'll check my inventory for a new C65 and replace it.

    But why didn't it boot with C65 disconnected?

    Thanks

    Bob

    #21 19 hours ago
    Quoted from oldschoolbob:

    I checked the resistance on pin 2 through pin 9. All showed about 7.4 K except pin 8 showed 4.9 K. Pin 8 is connected to C65.
    I checked resistance of C64 = 7.3 K. Then C65 = 4.9 K. Then C66 = 7.4 K. I pulled out the ground side of C65 and tried to boot - it still locks on at five flashes. With one leg of C65 disconnected I checked its resistance - it shows open (O.L). I checked the capacitance of C65 and it shows 0.49 nF.
    I'll check my inventory for a new C65 and replace it.
    But why didn't it boot with C65 disconnected?
    Thanks
    Bob

    C65 must not be the problem, but because you found an odd resistance reading there I'd say good chance the problem is on that PIA port. I just checked a board I have out and I get 7.75K resistance across ground to every digit drive port. Here is the kicker... when I short two digit drive PIA ports together i then get about 4.7K resistance across ground. I think you are going to find that U11 P6 is shorted to something else which is causing the PIA power on self test to fail.

    470pF = 0.47nF = 471 cap code. 0.49nF reading is close enough.

    #22 18 hours ago
    Quoted from barakandl:

    Here is the kicker... when I short two digit drive PIA ports together i then get about 4.7K resistance across ground.

    Thanks Andrew, After you said that I looked at the photo I posted before. Looks like there is something (solder?) between pin 33 and the trace from pin 8. Pin 33 is D0. And that's exactly what your talking about.

    I'm not in the shop right now but maybe after dinner I'll check there for a short. If that is a short it would explain why it still won't boot with C65 disconnected.

    Thanks

    Bob

    IMG_4204a (resized).jpg
    #23 18 hours ago
    Quoted from oldschoolbob:

    Looks like there is something (solder?) between pin 33 and the trace from pin 8.

    Good catch Bob. I think you found the issue..

    #24 17 hours ago

    I gobbled up dinner and when down to the shop. Yep, I had a short between 8 and 33. First I tried to clean it up with solder wick. That didn't work so I tried to scrape it off with a small pointed scribe. I finally got it cleared. Then I connected the power supply and got 6 flashes. I reconnected C65 and tested several more times. It works great. We brought another one back to life. Thanks guys - you're the greatest.

    All I need to do now is install the headers.

    Before

    S20200216_0003 (resized).jpg

    After

    S20200216_0004 (resized).jpg

    Thank you all.

    Bob

    Promoted items from the Pinside Marketplace
    From: $ 7,500.00
    $ 26.95
    Playfield - Protection
    ULEKstore
    $ 69.99
    Playfield - Toys/Add-ons
    Lighted Pinball Mods
    From: $ 7.99
    Playfield - Toys/Add-ons
    Medisinyl Mods
    $ 48.00
    Cabinet - Other
    ModFather Pinball Mods
    $ 234.99
    $ 78.99
    Playfield - Toys/Add-ons
    Lermods
    $ 100.00
    Playfield - Toys/Add-ons
    G-Money Mods
    $ 199.99
    Lighting - Led
    PinballBulbs
    $ 48.00
    Cabinet - Other
    ModFather Pinball Mods
    $ 10.00
    Playfield - Toys/Add-ons
    ModFather Pinball Mods
    $ 99.99
    Lighting - Other
    Lighted Pinball Mods
    $ 99.00
    Lighting - Under Cabinet
    Rock Custom Pinball
    $ 89.95
    Cabinet - Shooter Rods
    Super Skill Shot Shop
    $ 20.00
    Various Novelties
    GC Pinball
    $ 32.00
    $ 229.99
    Lighting - Led
    PinballBulbs
    $ 125.00
    Cabinet - Toppers
    Sparky Pinball
    $ 89.99
    $ 3.00
    $ 27.95
    Playfield - Toys/Add-ons
    ULEKstore
    $ 5.00
    Playfield - Decals
    Doc's Pinball Shop
    $ 99.99
    Lighting - Other
    Lighted Pinball Mods
    $ 79.00
    Cabinet - Armor And Blades
    PinGraffix Pinside Shop

    Hey there! Got a moment?

    Great to see you're enjoying Pinside! Did you know Pinside is able to run thanks to donations from our visitors? Please donate to Pinside, support the site and get anext to your username to show for it! Donate to Pinside